The Future of Security - Looking at how robots, AI, hacking, and more will make a difference

08/31/2016 - 16:57

Cole Mayer


If someone said a decade ago that soon nearly everyone in America would have a cell phone that could connect to the internet on a whim and with more computing power than a lunar lander, most would not believe them. While the future is hard to predict, it is possible to extrapolate what could be based on what we have. What does the future of security look like through today’s lens?

Robot Guards

Security guards are going to be a thing of the past. We already have a multitude of robot security guards rolling out.

Knightscope's K5 security bot came under fire recently for running over a toddler's foot and knocking him down while on duty patrolling the Stanford Shopping Center in California. Though there are some kinks that clearly need to be worked out, the bot has 360° cameras, thermal imaging, live streaming, autonomous charging, and even air quality sensors.

While bipedal robots like Asimo and Atlas have showcased their (limited) capabilities, they continue to advance and improve every year. Surmounting difficult terrain and stairs will give these machines an edge to other designs in the future. There is still a long way to go as you can see how slowly Atlas climbed stairs at the DARPA Robotics Challenge last year.

These robots are only going to get better and more efficient as technology progresses. The Law of Accelerating Returns will almost ensure this autonomous security guard revolution will come to pass one day.

AI and Cameras

What will make the aforementioned robots really efficacious is high-quality camera systems and software to analyze the incoming data. We already have off-site guards manning cameras, known as "virtual guards", which communicate with on-site guards and call for police assistance. When these security cameras are coupled with AI / deep learning algorithms to analyze and monitor for safety, we will greatly surpass human capacity in this field.

“While a security officer might miss a person sneaking into a poorly lit facility, a camera backed with intelligent video analytics is designed to catch a flash on the screen and recognize it as a potential threat...” - Dr. Mahesh Saptharishi

He further states in an article for Wired, “Or it will spot a person loitering at the perimeter of a schoolyard and alert on-the-ground security officials to investigate and take action if necessary, all without missing a beat and keeping close watch on the many cameras and locations.”

The on-the-ground official would, in our scenario, be the automaton security guard. With the security camera system and the onboard camera system for the bot, the AI would enable the two technologies to seamlessly work together.

Drones

This can be expanded even further to incorporating security drones equipped with high resolution cameras, thermal imaging, and much more. The military is already incorporating tethered drones for security purposes. The addition of the drone can add further insight the other systems are not capable of and is incredibly able to cover a wide area quickly. Given drones short battery life, a larger and more efficient battery would be required or incorporating autonomous charging.

We have already seen drone enthusiasts chase down boat thieves who were then caught by the authorities! This will become normal for autonomous security systems that utilize drones.

There is even a company in Japan that specializes in home drone security systems. While these security drone systems are currently quite uncommon, they will likely become adopted as economies of scale comes into play, the technology gets better and becomes cheaper.

NFC Chips - Near Field Communication

Controlled access for the human workers at a site will also change. Imagine doing away with keycards and instead using an implant smaller than a grain of rice that could verify identities at checkpoints, provide medical info, be used for money transfer, and many other possibilities.

A man last year used a near field communication chip implanted into his hand to send Bitcoin to another wallet. This type of biohacking has been happening for years and the aptly named, Dangerous Things, is at the forefront of the movement. The founder, Amal Graafstra, first embedded the chip into his hand more than a decade ago solely to open doors.

Even though these chips inside the human body are deemed safe, security is always an issue. The various protocols will need to have reliable encryption and be open source. Hacking is something everyone needs to take seriously and ensure they try their best to stay protected. Obviously, many of the key components are out the users' abilities and up to the manufacturers to ensure the vanguard of safety. NFC hacking has been a standard topic at Defcon and here is a brief guide, "NFC Hacking: The Easy Way" (PDF). The community needs to be able to verify these systems ensuring no exploits that exist that could cause loss of money, health info, etc. Multi-factor authentication would probably be necessary to ensure the highest security capability. Nonetheless, NFC could even be used as an authentication device.

Regardless, Edward Snowden has come out in a white paper (PDF) against the use of NFC. He wants to remove the feature for those dealing with sensitive materials (ie. journalists exposing leaks), "...(to) prevent exploitation of the radio without leaving a clear signature that can be detected by an adversary."

While the concept of implanting chips into our bodies does seem like an Orwellian scenario that would have privacy advocates up in arms, it is another trend likely to come about. There is a Biblical reference to this type of concept named the Mark of the Beast, that would have religious groups dissenting to any sort of required implanted chips as well.

Skynet and Hacking

There are potential downsides. In a Terminator Skynet scenario, the AI becomes sentient and decides it does not want to work for humans anymore - or decides to eliminate them. The elimination may be due solely to the task the system was given and it simply extrapolates the task to such a degree, it overruns the entire species on Earth (e.g. an AI paper clip factory). To learn more about this, watch these enlightening videos from Nick Bostrom and Pedro Domingos.

While this scenario coming to pass is probably farfetched, having AI as security guards, drone systems, and masters of surveillance without any human oversight could be unwise. Having well-written algorithms and a kill switch to such systems could help prevent a potential robot uprising. That is of course assuming the AI doesn't reprogram itself and disable such kill switch without human knowledge.

Needless to say, a more pertinent threat is hacking. As we know, anything can essentially be hacked. Even the NSA. While cyber security is continually advancing, not all organizations are following suit and hackers continue to exploit this. Motherboard has an ongoing feed about this given the continuous hacks happening, "Another Day Another Hack."

Hackers could, in theory, get root access to the system controlling the drones, robot guards, etc., effectively shutting them off or spoofing the camera feed to suit their purpose. Human oversight could be an additional defense against this, but is definitely not a failsafe.

As technology continues to accelerate from both the hardware and software perspectives, these threats can all be greatly mitigated. Having open source systems with established encryption protocols (and NO back doors) will enable greatly advanced future security operations.